• TalkBass has been independent since 1998. Add your voice.
    Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
    Join freeLog in
    Want zero display ads or expanded classifieds tools? Compare plans.

Op-amp buffer questions

So I'm working on building a simple buffer circuit for one of my basses. I've seen the JFET based ones floating around and I get those, but I'd like to try an op-amp based one. Mainly for the much lower output impedance and the lower parts count is nice too, as space is limited in my control cavity.

My questions are:
1. Power supply. I only really have space for one 9-volt battery, so that limits what I can do. Do I want to do it dual or single rail? Can I do dual rail with just one battery? And how do I set up the power supply? Most schematics don't go over setting up a power supply for either in much detail, so I'm kind of in the dark on this.

2. Controls. Instead of a tone knob, I just use an on-off-on SPDT with two different capacitor values; I also use a killswitch in place of volume. I'm pretty sure these can go after the buffer without any issues, but I'd like to know if I'm missing anything that might make a difference.

I'm sorry if these questions seem kind of out there or uninformed, but I'm still a little new to active electronics. Much thanks to anyone who can help.
 
You might be able to pull something from this one. It uses a single 9V:
sr480_preamp.jpg


Link Removed's a pdf of an ASAT bass schematic from the [Invalid or Expired Link Removed].
 
Power supply: To do dual rail with a single battery, you could create a virtual ground, commonly done with voltage divider (R14 & R15 in the schematic above). For better performance, have a look at Texas Instruments’ TLE2426 'Rail Splitter' Precision Virtual Ground.

Or, for even more headroom, go with a charge pump voltage converter to get a -9V rail. Link Removed or [Invalid or Expired Link Removed]
 
Here is an extremely simple opamp buffer schematic: Invalid Link Removed
2 caps, 2 resistors and 1 IC.

That's the basic schematic, but I'm not sure that I would use the NE5534 for a battery powered buffer. The datasheet says that it's stabilized for gains above three, it draws 4 mA, has a 5-Volt output swing when powered at 9 V, and will have an output offset upwards of 0.5 V in the schematic shown. There may be more to the circuit than shown in the schematic, e.g., the components for unity gain compensation of the 5534.

As for the parts count, a JFET source follower can have as few as 4 parts, as there's no need for an artificial ground. I'm partial to the 2N5486 because its output bias voltage is somewhere around 1/2 of 9 Volts when used as a buffer.
 
Thanks for the responses so far. The G&L diagram was handy, though it's a different chip than the ones I have. I like the charge pump idea, I'll have to remember that for some designs that need more headroom. The chip I have on hand at the moment (some Fry's brand thing, picked it up just because it was there) has a pinout for V+ and one for ground, so I'd assume it only needs the positive rail.

fdeck, the source follower is what I'm looking at making at the moment, though I think the output impedance is a bit higher than I'd like for most JFETs. Also, don't I need to bias the input for them?
 
At my website, I've got a "quick and dirty" preamp design which might be worth looking at. You can get away without biasing, if the JFET happens to have the right cutoff voltage. The 2N5486 is around 4 V give or take. There is some natural variation in cutoff voltage from one JFET to the next, even of the same type. You can deal with that variation by:

1. Buying a dozen JFETs and picking the nicest one
2. Adding a bias adjustment
3. Living with the variation

The third option is what I typically go with.

As shown in my schematic, the output impedance is 1 k Ohms. In reality you could probably get away with even higher output impedance. I'm figuring that if driving a typical instrument cable, 20 feet at 50 pF / foot, 10k output impedance would result in a cutoff frequency of 15 kHz.
 
I've seen your "quick and dirty" pre, it's one of the ones I was looking at making. I figure I'd go with setting up an input bias for the JFET, as it wouldn't take up that much more space. From what I've picked up so far, it would just take an extra resistor from the input to the power source. Again, thanks for the help, I've still got quite a bit to learn about this stuff.
 
I've seen your "quick and dirty" pre, it's one of the ones I was looking at making. I figure I'd go with setting up an input bias for the JFET, as it wouldn't take up that much more space. From what I've picked up so far, it would just take an extra resistor from the input to the power source. Again, thanks for the help, I've still got quite a bit to learn about this stuff.

If you use a circuit similar to the Alembic Stratoblaster, you don't need to bias the JFET because it's self biasing.

It's a nice sounding buffer too.
 
The Bias or operating point for a simply jfet buffer is just at 1/2 of the supply voltage.

You can eliminate alot of problems by simply just using a voltage divider used for Virtual ground with single supply opamp circuits.

its simple and unlike various methods it can cause less noise and will always be at 1/2 supply even if the battery voltage falls...so there you go "self bias" at any supply voltage.

plus you can change the input impedance to any value with a single resistor and bias will always remain at 1/2 supply via the divider

[Invalid or Expired Link Removed]
 
Since this is getting pretty in-depth (and thankfully so) - have any on-board buffers or preamps ever been designed in a push-pull configuration, in the sense that there are two FETs or op-amps, each handling half the waveform?

You could do something with back to back P- and N-type JFETs. Part of the hassle is that 9 V is a pretty meager budget when you're stacking multiple voltage drops.
 
Since this is getting pretty in-depth (and thankfully so) - have any on-board buffers or preamps ever been designed in a push-pull configuration, in the sense that there are two FETs or op-amps, each handling half the waveform?

Yes and they are called Jfet class AB operational amplifiers.
Common use is for Low Voltage Rail to Rail opamps.

So the whole package would just be a standard opamp, and you would use it just like any other opamp.

But im sure your looking for a discrete build for some sorta cool mojo....otherwise completely unneeded in a simple buffer.
That design is suited for driving low impedance sources or high capacitance loads.

or if you look at some designs used in the 70's and 80's to drive reverb tanks, which uses a opamp to drive a pair of transistors . You could modify it to to your needs and update whatever ancient opamp they are using...again uneeded really more useful for high current...which can be replaced by hundreds of 1watt to 5 watt chip amps
 
What I want to know is why an 062 OpAmp? :confused:

What would you recommend? I'm partial to the rail-to-rail TLC220x and TLC226x for 9-V operation, due to tolerably low noise and low operating current. Chips with much lower noise tend to get pretty battery-hungry, a reason why I continue to fall back on discrete designs.
 
What would you recommend? I'm partial to the rail-to-rail TLC220x and TLC226x for 9-V operation, due to tolerably low noise and low operating current. Chips with much lower noise tend to get pretty battery-hungry, a reason why I continue to fall back on discrete designs.

I've built several opamp based preamps with the LT1351/2/3 chips with great results. Linear Tech will give you 2 free samples of each chip and they will accept low quantity orders. They are only $4 a piece and it's a steal for the results I get.